Project Summary The human hand is critically important for performance of many activities of daily living (ADL). Impaired neural control of the hand due to traumatic injury or neurological disease could significantly diminish patients? overall quality of life. Despite advances in our understanding of how the central nervous system learns to control arm movements from repetitive exercises, little is known about the role of force repetition in learning predictive and corrective control of hand-object interactions. The proposed project tests the hypotheses that (1) motor repetition of simple, isometric finger forces improves hand control in complex, untrained manipulation tasks, and (2) the extent of the improvement in the control of manipulation is modulated by the type of motor repetition, as well as by aging. We propose a series of experiments on healthy young (age 18-40) and older (age 60-75) human participants, who will perform one of four types of repetitive exercise of the hand. The motor behavior and structure of the repetition differ. There are two force-based repetition trainings, one of which focuses on producing accurate ballistic force pulses whereas the other focuses on sustained exertion of forces guided by visual feedback. As comparisons, there are two motion-based repetition trainings which have either ballistic or sustained hand movements. After training, we will systematically examine the effect of each repetition type on novel (untrained) manipulation tasks that are generated by robotic devices. The first generalization task requires participants to predictively compensate for unfamiliar viscous dynamics during object lifting. We will quantify how these repetition training paradigms differentially shape the adaptation of predictive hand control by evaluating participant?s performance in the novel dynamics. The second task assess the participants? ability to use feedback-based corrective control for stabilizing a simulated virtual pendulum. We will quantify the extent to which these four training paradigms modulate the speed and accuracy of the reactive control. The effect of repetition type and aging will be examined. The long-term objective of the proposed research is to gain insights into the neural mechanisms underlying the effectiveness of different repetition training types, and their generalization to novel learning and control tasks. These findings will inform future studies investigating the plasticity of neural mechanisms dictating the relation between motor repetition and recovery of sensorimotor function. Furthermore, the results of the proposed studies will provide the scientific basis to 1) improve the efficacy of sensorimotor rehabilitation protocols, 2) inspire design of low-cost rehabilitation devices, and 3) determine objective metrics to quantify successful recovery of sensorimotor function for human hands.